hendrickcars.com — AI Site Grade

Analysis

I have enough information now. Let me compile the audit.

---

The site serves full rich HTML to anthropic-ai while returning 403 Access Denied to every other AI crawler and even to standard browsers via Akamai.

Crawler Access

The site operates a two-tier access system. Akamai sits in front and blocks all traffic from browsers, GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, PerplexityBot, Google-Extended, Bytespider, and Applebot-Extended — all return 403 with a 372-byte "Access Denied" page. The anthropic-ai user-agent bypasses Akamai entirely, hitting a separate nginx origin that returns 200 with ~692KB of full HTML. The robots.txt (accessible only via anthropic-ai) explicitly disallows GPTBot, OAI-SearchBot, ChatGPT-User, Claude-User, Claude-SearchBot, and PerplexityBot from /api/, /apis/, and static asset paths — but does not block them from the main site. The block is enforced at the CDN layer, not the robots.txt layer. No llms.txt exists (returns 403 to all agents).

Content & Schema

The homepage served to anthropic-ai is a Dealer.com (DDC) platform build — a JS-heavy single-page application shell with 30+ brand CSS classes (Chevrolet, BMW, Porsche, Lexus, etc.). The visible <title> is "HendrickCars.com - New & Used Car Dealers in Charlotte, Charleston & Across the U.S." The page is ~692KB of HTML but is almost entirely JavaScript bootstrap code, <link> tags, and <script> references. No JSON-LD schema was extractable from the homepage via standard fetch (which gets 403). Subpages like /wlp-car-dealership-in-charlotte-nc.htm serve 267KB of similar JS-heavy HTML to anthropic-ai. The sitemap.xml (accessible only to anthropic-ai) lists hundreds of URLs with lastmod dates as far out as 2026-05-30 — a suspiciously future-dated timestamp suggesting automated generation without actual content freshness tracking.

Cold-Knowledge Gap

The LLM knows Hendrick Cars as "a network of auto dealerships primarily selling GM brands" tied to NASCAR owner Rick Hendrick, with a "strong presence in the Carolinas." It also recalls a 2023-2024 lawsuit over deceptive pricing and add-on fees. The site itself never surfaces this lawsuit or any response to it. The cold knowledge is richer on the brand's motorsports heritage and legal history than anything the site's homepage actually communicates to AI crawlers — because only one bot (anthropic-ai) can even read it.

External Signals

No external search results returned for site:hendrickcars.com or brand-specific queries — the site appears to have zero indexed pages in standard web search. The Wayback Machine has no snapshots of the domain. DNS records show the site runs on Akamai (CDN) with GoDaddy email (MX: smtp.secureserver.net). The domain uses DNSimple for nameservers. The only discoverable external signal is the cold LLM knowledge of the Hendrick brand's NASCAR connection and the pricing lawsuit — neither of which the site addresses.

Surprising Architecture

This is a dealership group running a public-facing website that blocks virtually all organic discovery. The Akamai WAF configuration appears to be a deliberate access-control strategy, not a misconfiguration: it consistently blocks browsers (403) while allowing exactly one AI crawler through to a separate nginx backend. Standard /inventory and /about paths return 404 even to anthropic-ai. The site is effectively invisible to Google, Bing, and every major AI crawler except Claude's training crawler. The sitemap's 2026 lastmod dates suggest the platform auto-generates timestamps without reflecting real content updates.

Findings

1. Akamai blocks all AI crawlers except anthropic-ai High

   The site's Akamai CDN returns 403 Access Denied to GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, PerplexityBot, Google-Extended, Bytespider, Applebot-Extended, and standard browsers. Only the anthropic-ai user-agent bypasses Akamai and receives a 200 response from a separate nginx origin.

   What to change: Reconfigure Akamai WAF to allow standard AI crawlers (GPTBot, OAI-SearchBot, etc.) and browser traffic, or implement a unified access policy that does not discriminate between user-agents.

2. robots.txt inaccessible to most bots High

   The robots.txt file returns 403 to all user-agents except anthropic-ai, preventing standard crawlers from reading crawl directives. Even when accessed by anthropic-ai, the file disallows GPTBot, OAI-SearchBot, ChatGPT-User, Claude-User, Claude-SearchBot, and PerplexityBot from certain paths, but the actual block is enforced at the CDN layer.

   What to change: Make robots.txt publicly accessible and ensure it accurately reflects crawl permissions for all bots.

3. No llms.txt file Medium

   The site does not provide an llms.txt file; requests return 403. This file would help AI crawlers discover key pages and understand site structure.

   What to change: Create an llms.txt file listing important pages (inventory, about, specials) to guide AI crawlers.

4. Homepage is a JS-heavy shell with no extractable JSON-LD High

   The homepage served to anthropic-ai is ~692KB of HTML but consists almost entirely of JavaScript bootstrap code, CSS links, and script references. No JSON-LD structured data was extractable, which limits AI understanding of the site's content and entities.

   What to change: Implement server-side rendering or static HTML for key content, and include JSON-LD structured data (e.g., AutoDealer, VehicleOffer) in the initial HTML.

5. Inventory and about pages return 404 even to allowed bot High

   Standard paths like /inventory and /about return 404 to anthropic-ai, meaning the site's core content is inaccessible to AI crawlers. Only a few pages like /wlp-car-dealership-in-charlotte-nc.htm are accessible.

   What to change: Ensure inventory and about pages return 200 with meaningful content to AI crawlers, or provide alternative accessible URLs.

6. Sitemap contains future lastmod dates (2026) Medium

   The sitemap.xml, accessible only to anthropic-ai, lists hundreds of URLs with lastmod dates as far out as 2026-05-30, indicating automated generation without actual content freshness tracking. This misleads crawlers about content recency.

   What to change: Update sitemap generation to use accurate lastmod dates reflecting actual content updates.

7. Zero pages indexed in web search High

   No search results returned for site:hendrickcars.com or brand-specific queries, indicating the site has no indexed pages in Google or Bing. This is consistent with the blanket blocking of standard crawlers.

   What to change: Allow Googlebot and Bingbot through Akamai to enable indexing, and ensure content is crawlable.

8. No Wayback Machine snapshots Low

   The domain has no snapshots in the Wayback Machine, suggesting it has never been publicly accessible to archiving bots, further confirming the access restrictions.

9. Cold knowledge reveals lawsuit not addressed on site Medium

   LLM cold knowledge recalls a 2023-2024 lawsuit over deceptive pricing and add-on fees, but the site does not address or reference this legal issue, creating a potential reputational gap for AI-generated summaries.

   What to change: Consider adding a page or statement addressing the lawsuit to provide context for AI crawlers.

10. No external backlinks or signals detected Medium

    Web searches for Hendrick Cars dealerships and reviews returned no results, indicating a lack of external signals that could boost AI visibility.

    What to change: Build external backlinks through local SEO, partnerships, and listings to improve discoverability.

What's working

• anthropic-ai receives full rich HTML — The site serves a complete ~692KB HTML page to the anthropic-ai crawler, including brand classes and metadata, enabling Claude to parse the site's content.
• Sitemap accessible to anthropic-ai — The sitemap.xml is served to anthropic-ai, listing hundreds of URLs that could guide crawling if other bots were allowed.
• robots.txt allows main site for listed bots — The robots.txt file (when accessible) does not disallow GPTBot, OAI-SearchBot, etc. from the main site, only from /api/ and static paths, which is a reasonable configuration.
• Strong brand recognition in LLM cold knowledge — LLMs know Hendrick Cars as a major dealership network with NASCAR ties, providing baseline awareness that the site can build upon.
• Special pages like monthly specials and location pages accessible — Pages like /monthly-specials-details.htm and /wlp-car-dealership-in-charlotte-nc.htm return 200 to anthropic-ai, providing some content for AI consumption.